Elsevier Editorial System(tm) for Journal of Food Engineering Manuscript Draft Manuscript Number: JFOODENG-D-14-00374R2 Title: Effect of water activity in tortilla and its relationship on the acrylamide content after frying Article Type: Research Article Keywords: Acrylamide; Tortilla chips; Minimum integral entropy Corresponding Author: Dr. Ricardo Salazar, Ph.D Corresponding Author's Institution: Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional First Author: Rosa M Delgado, Ph.D Order of Authors: Rosa M Delgado, Ph.D; Gabriel Luna-Bárcenas, Ph.D; Gerónimo Arámbula-Villa, Ph.D; Ebner Azuara, Ph.D; Patricia López-Pérea, Ph.D; Ricardo Salazar, Ph.D Abstract: The objective of this study was to relate the tortilla minimum integral desorption entropy with acrylamide content during processing of tortilla chips. Tortilla pieces were stored at 30 ˚C at aw of 0.11-0.84 for 4 days and fried later in soybean oil at 180 ˚C for 25 s. The lowest acrylamide content was observed in tortilla chips made of non-stored tortilla (aw=0.98) as well as in those prepared from tortilla stored in the minimum integral entropy (aw=0.53). In addition, the color and texture values were similar in both cases. These results suggest that the reduction of the acrylamide content during processing of tortilla chips and other tortilla based foods thermally processed might be modified by factors such as moisture content, aw, and the physical state of water in the tortilla. Thus, the minimum integral entropy showed to be a reliable indicator to establish the most appropriate moisture conditions to obtain tortilla chips with reduced level of acrylamide when tortilla is dehydrated.
Highlights
Water activity affects the acrylamide content in tortilla chips.
Minimum integral entropy is a reliable indicator to establish the most appropriate
moisture conditions to get a reduction of acrylamide in tortillas chips.
Tortilla dehydration to the minimum integral entropy reduces the acrylamide
content.
Products obtained from stored tortilla showed an acceptable texture and appearance.
*Highlights (for review)
1
Effect of water activity in tortilla and its relationship on the acrylamide 1
content after frying 2
Rosa M. Delgadoa, Gabriel Luna-Bárcenasb, Gerónimo Arámbula-Villab, Ebner Azuarac, 3
Patricia López-Peréad and Ricardo Salazarb*. 4
a Instituto de la Grasa, Consejo Superior de Investigaciones Científicas, Avenida Padre 5
García Tejero 4, 41012, Seville, Spain 6
b Centro de Investigación y de Estudios Avanzados del I.P.N., Unidad Querétaro, 7
Libramiento Norponiente # 2000, Fraccionamiento Real de Juriquilla, 76230, Querétaro, 8
Querétaro, México 9
c Instituto de Ciencias Básicas, Universidad Veracruzana. Av. Dr. Rafael Sánchez 10
Altamirano s/n, Col. Industrial Animas, Carretera Xalapa-Las Trancas Km. 3.5, 91192, 11
Xalapa, Veracruz, México 12
d Facultad de Ciencias Agrícolas, Campus El Cerrillo, Universidad Autónoma del Estado 13
de México. Piedras Blancas, Carretera Toluca - Ixtlahuaca Km. 15.5, 5020, Toluca, Estado 14
de México, México. 15
16
* Corresponding author. Tel. +52 442 211 9900; fax: +52 442 211 9938. E-mail address: 17
[email protected] (Ricardo Salazar) 18
19
*ManuscriptClick here to view linked References
2
Abstract 20
The objective of this study was to relate the tortilla minimum integral desorption entropy 21
with acrylamide content during processing of tortilla chips. Tortilla pieces were stored at 30 22
˚C at aw of 0.11-0.84 for 4 days and fried later in soybean oil at 180 ˚C for 25 s. The lowest 23
acrylamide content was observed in tortilla chips made of non-stored tortilla (aw=0.98) as 24
well as in those prepared from tortilla stored in the minimum integral entropy (aw=0.53). In 25
addition, the color and texture values were similar in both cases. These results suggest that 26
the reduction of the acrylamide content during processing of tortilla chips and other tortilla 27
based foods thermally processed might be modified by factors such as moisture content, aw, 28
and the physical state of water in the tortilla. Thus, the minimum integral entropy showed 29
to be a reliable indicator to establish the most appropriate moisture conditions to obtain 30
tortilla chips with reduced level of acrylamide when tortilla is dehydrated. 31
32
Keyword: Acrylamide, Tortilla chips, Minimum integral entropy 33
34
Chemical compounds studied in this article 35
Acrylamide (PubChem CID: 6579) 36
3
1. Introduction 37
The Maillard reaction is the main responsible of the generation of flavor and color in 38
thermally processed foods (Ames, 1990). However, this reaction has been also linked to the 39
acrylamide formation, specifically with the presence of carbonyl compounds with groups 40
capable of forming a Schiff Base with the amino acid asparagine (Hidalgo et al. 2009; 41
Mottram et al. 2002; Stadler et al. 2002). Acrylamide, a neurotoxic compound (Spencer & 42
Schaumburg, 1974) and probable human carcinogen (IARC, 1994), has attracted the 43
attention of the scientific community in recent years due to it has been found in high 44
concentrations in thermally processed foods (Tareke et al. 2002). 45
The effect of water activity (aw) on acrylamide formation has been widely studied. 46
Depending on the water activity range studied, positive or negative correlations between 47
water activity and acrylamide content have been observed. De Vleeschouwer et al. (2007) 48
showed very slight promoting effect of water activity (0.34≤aw≥0.92) on acrylamide 49
amount in an asparagine–glucose model system. Similar results were reported by De 50
Vleeschouwer et al. (2008) (0.88≤aw≥0.99). The same tendency was observed in a potato 51
model system (0.11≤aw≥0.97) where the acrylamide content was rather dependent upon the 52
moisture content than upon the water activity (Mestadgh et al. 2006). However, an opposite 53
trend was found in aqueous glycerol asparagine model system (0.33≤aw≥0.71) (Hedegaard 54
et al. 2007) and plantain paste (0.43≤aw≥0.97) (Bassama et al. 2011), revealing a decrease 55
in acrylamide formation with increasing water activity. 56
The above-mentioned studies have in common that the food model showed a sorption 57
isotherm typical of sugar-rich products (Mathlouthi, 2001) where a small variation in 58
4
moisture content corresponds to a large variation in aw for water activities lower than 0.9. 59
For water activities higher than 0.9, a steep increase in the corresponding moisture content 60
can be observed (De Vleeschouwer et al. 2007). In contrast, most foods are a mixture of 61
different compounds (for example, carbohydrate, protein, fat, minerals, vitamins, salts and 62
others). Hence, real foods tend to have a sigmoidal moisture sorption isotherm due to the 63
presence of solutes with very strong affinity for water that could maintain a larger amount 64
of bound moisture at lower humidity than solutes with less affinity. 65
Water activity influences on the stability and chemical reaction rates in foods (Labuza et al. 66
1972). Thus the mobility of acrylamide precursors and as a consequence, the acrylamide 67
formation is affected by the physical state of the water sorbed during thermal processing. In 68
this context, it is expected that the final acrylamide content will be mainly affected by the 69
moisture content (monolayer value) where strong bonds between the water (adsorbate) and 70
the food (adsorbent) occur. 71
In an attempt to understand the influence of water activity of real foods on acrylamide 72
content, the level of acrylamide was compared in tortilla chips prepared from tortilla 73
equilibrated at different water activities, including that corresponding to the monolayer 74
value. The monolayer was calculated from tortilla desorption isotherms by using both, 75
GAB equation and the minimum integral entropy criteria. 76
2. Materials and methods 77
2.1 Materials 78
Labeled [2,2,3-2H3]acrylamide was purchased from Sigma–Aldrich (St. Louis, MO). All 79
other chemicals were analytical grade and purchased from Sigma (St. Louis, MO) or Merck 80
5
(Darmstadt, Germany). Soybean oil was obtained from local supermarkets in Querétaro, 81
México. 82
2.2. Elaboration of nixtamalized corn flour and tortilla chips. 83
Nixtamalized corn flour was prepared with commercial corn variety named Pioneer 30P16 84
and commercial lime (Ca(OH)2) (El Topo, Monterrey, N.L. Mexico), commonly used in the 85
tortilla industry. This flour was prepared by cooking (8 kg) of whole corn kernels in a 86
solution of 16 liters of water with 80 g of Ca(OH)2, corresponding to 1.0 g/100 g of lime 87
relative to the corn weight used. The corn was boiled in an aluminum pan for 25 min and 88
steeped for 16 h at room temperature (22 ± 1°C). The steep liquor was removed. The 89
cooked corn was washed with 16 liters of water, then ground into corn dough (FUMASA, 90
M100, Querétaro, México), and finally dehydrated using a flash type dryer (Cinvestav-AV, 91
M2000, Querétaro, México). The drying conditions were adjusted to have 250°C inlet air 92
temperature and 90°C to the exhaust air to avoid burning the material. Before storage, the 93
nixtamalized corn flour was milled using a hammer mill (PULVEX 200, México DF) 94
equipped with a 0.5 mm screen. 95
For tortilla chips elaboration, nixtamalized corn flour (100 g) was rehydrated with enough 96
water (118 mL) to provide fresh dough with proper consistency to make tortillas. The 97
dough was shaped into thin disks (11cm diameter and 1.0 mm thickness) using a 98
commercial tortilla roll machine (Casa Herrera, México, D.F.). The dough shaped into 99
tortillas were cooked on both sides for around 1.0 min by using an iron hot plate (270 ± 10 100
°C) named “comal”. The resulting tortillas were cut into circular pieces with an average 101
area of 10 cm2 and its aw and pH value (AACC International, 2010) determined. Two set of 102
6
tortilla pieces were obtained. The first set was fried immediately in soybean oil at 180°C 103
for 25 s (control). The second set was used in the construction of desorption isotherms. 104
After frying, tortilla chips were cooled on a paper towel to remove superficial oil and the 105
color, fracture force and acrylamide content determined. 106
2.3. Water desorption isotherms 107
Water desorption isotherms were determined using a static equilibrium method. Tortilla 108
samples (∼4.0 g) (aw =0.98) were weighed in triplicate into standard weighing dishes with a 109
circular section on the bottom. Samples were placed in separate desiccators containing 110
saturated salt slurries in the range of water activity from 0.11 to 0.85 using the aw values 111
reported by Labuza et al. (1985). The samples were held at 25, 30, and 35 °C until 112
equilibrium was reached. Values of water activity were generated using equations reported 113
in the same paper. A small amount of toluene was placed in each desiccator to prevent the 114
growth of fungi. The equilibrium condition was attained within 7-10 days, when the 115
differences among two consecutive weights were within 0.001 g. The dry matter content of 116
the tortilla was determined by the drying in a conventional oven at 105 °C for 24 h. The 117
Guggenheim-Anderson-De Boer (GAB) equation was used in modeling water sorption 118
(Quirijns et al. 2005): 119
M MCka1 ka1 ka Cka1
Where aw is water activity; M is water content of the sample on dry basis; M0 is the 120
monolayer water content; C is the Guggenheim constant, given by C= c’exp (hm-hn)/RT; 121
where c’ is the equation constant; hm is the heat of sorption of the first layer; hn is the heat 122
of sorption of the multilayer; R is the gas constant; T is the absolute temperature; and k is 123
7
the constant correcting properties of multilayer molecules with respect to bulk liquid, and 124
given by k= k’ exp (h1-hn)/RT; where k’ is the equation constant; h1 is the heat of 125
condensation of pure water. The parameters values of GAB equation (M0, C and k) were 126
estimated by fitting the mathematical model to the experimental data, using non-linear 127
regression with the Kaleidagraph 4.0 package (Synergy Software, Perkiomen, USA). 128
Goodness of fit was evaluated using the average of the relative percentage difference 129
between the experimental and predicted values of the moisture content or mean relative 130
deviation modulus (P) defined by the following equation: 131
P% 100N |Me Mc|Me
2
Where Mei is the moisture content at observation i; Mci is the predicted moisture content at 132
that observations; and N is the number of observations. It is generally assumed that a good 133
fit is obtained when P < 10% (Lomauro et al. 1985). 134
2.4 Determination of minimum desorption integral entropy 135
The determination of the integral (enthalpy and entropy) thermodynamic properties, and 136
the water activity-temperature conditions where the tortilla minimum integral entropy 137
occurred, considered as the point of maximum storage stability (monolayer value), was 138
established as indicated by Pascual- Pineda et al. (2013) and Vigano et al. (2012). These 139
authors have provided a thorough description of the procedure followed and equations used 140
for this purpose. Briefly, the integral enthalpy changes (∆Hint)T (J/mol) at the water-tortilla 141
interface and, at different stages of the adsorption process, were determined using the 142
equation of Othmer (Othmer, 1940). 143
8
dlnPdlnP
HTH!T3
Where: the desorbed substance is water; #$(Pa) is the vapor pressure of water over the 144
adsorbent; #$ (Pa) is the vapor pressure of pure water at the temperature of sorption; %$& 145
(J/mol) is the integral molar heat of sorption, and %$& (J/mol) is the heat of 146
condensation of pure water. Since all these terms are temperature-dependent, the equation 147
can be integrated: 148
'(#$ )%$&%$*&+, '( #$ -4
Where: A is the adsorption constant, and Ф (J/mol) is the pressure of diffusion or surface 149
potential. A plot of '(#$versus '( #$gives a straight line if the ratio %$&/%$& is 150
constant within the range of temperatures used. 151
The molar integral enthalpy (∆Hint)T can be calculated using equation (5), at a constant 152
pressure of diffusion( Nunes & Ronstein, 1991): 153
∆%1234 5%$&%$*& 16,%$*&5
8 9:;<9: =&>:;>$ ? @A'(BC:D
6
Where: 9:; (J/mol) is the chemical potential of the pure adsorbent; 9: (J/mol) is the 154
chemical potential of the adsorbent participating on the condensed phase; >:; (g/mol) is 155
the molecular weight of the adsorbent, and >$ (g/mol) is the molecular weight of the 156
water. 157
9
By calculating %$&/%$& from equation (4) and substituting it into equation (5) it 158
becomes possible to calculate the integral enthalpy at different temperatures, provided that 159
a good means of estimating is available, such as that proposed by Wexler, (1976): 160
%$& J/mol-K = 6.15 x 104 – 94.14 T + 17.74 x 10-2T2 – 2.03 x 10-4T3 (7) 161
Using the values obtained for (∆Hint)T changes, the molar integral entropy (∆Sint)T can be 162
estimated using the following equation : 163
(∆Sint)T=S1-SL= -(∆Hint)TT
-Rlnaw 8
where: F F/G(J/mol K) is the integral entropy of water desorbed in the foodstuff; S 164
(J/mol K) is the total entropy of water desorbed in the foodstuff; N1 is the moles of water 165
desorbed in the foodstuff, and SL (J/mol K) is the molar entropy of pure liquid water in 166
equilibrium with vapor. 167
2.5. Storage of tortilla before frying 168
Since water retention capacity in tortillas is poor, the moisture loss in tortilla after 169
processing is often caused by starch retrogradation. Taking advantage of this phenomenon, 170
tortilla samples were left at room temperature (25 °C) during 4 h to reduce their aw from 171
0.98 to 0.85 and therefore, reducing the moisture content and preventing spoilage reactions 172
during the time that the samples reached the desired aw. The tortilla pH was monitored in 173
order to confirm that acrylamide levels were not affected by pH changes during storage. 174
Duplicate samples containing ca. 5 g of tortilla were placed in desiccators containing the 175
same saturated salt slurries referred in section 2.3 at 30°C. Tortilla samples were withdrawn 176
at day 5 to prepare tortilla chips. 177
Hvo(T)
10
2.6. Color determination in tortilla chip samples 178
Color changes were determined using a colorimeter MiniScan XE, model 45/0-L (Hunter 179
Associates Laboratory, 11491 Sunset Hill Rd., Reston, Va., U.S.A.). Total color differences 180
(∆E) at the different periods of time were calculated from the determined CIELAB L* a* 181
b* values according to Hunter (1973): ∆E= [(∆L* ) 2 + (∆a*) 2 + (∆b*) 2]½; where L* = 182
brightness or lightness (100 = perfect white, to 0 = black); a* = greenness/redness [negative 183
(green) to positive (red)]; b* = yellowness/ blueness [negative (blue) to positive (yellow)]; 184
∆L* , ∆a*, and ∆b* = absolute differences of the values between the reference tile (white 185
porcelain) and sample values; ∆E = total difference between reference and sample color. 186
The reference values (calibration) were: L* = 92.22, a* = -0.82 and b* = 0.62. 187
2.7. Texture determination in tortilla chips 188
The fracture force of the tortilla chips was evaluated using the Texture Analyzer TA-XT2 189
(Texture Technologies Corp., N. Y.). Fracture force was evaluated in freshly prepared 190
samples. The test was carried out using a 2.03 mm diameter stainless-steel probe and a 191
platform accessory with a hollow cylindrical base with 33.5 and 10 mm external and 192
internal diameters, respectively. The probe traveled at a velocity of 10 mm/s until it cracked 193
the sample (distance 6 mm). 194
2.8. Acrylamide determination in tortilla chips 195
Acrylamide was analyzed as the stable 2-bromopropenamide derivative by gas 196
chromatography–mass spectrometry (GC–MS) using a combination of the methods of 197
Andrawes et al. (1987) and Castle et al. (1991) as described previously (Salazar et al. 198
2012). Tortilla chips were ground in a mortar lab and powdered samples (∼0.8 g) were 199
11
successively weighed in centrifugal tubes, spiked with 20 µL of internal standard solution 200
(0.5 mg/mL of deuterium-labeled [2,2,3-2H3]acrylamide in acetonitrile), and stirred with 8 201
mL of distilled water and 10 mL of n-hexane at room temperature for 5 min. After 202
centrifugation at 2000 x g for 10 min, organic phases were removed. Co-extractives from 203
supernatants were precipitated with 30 µL of carrez I and 30 µL of carrez II solutions. 204
Later, supernatants were centrifuged at 2000 x g for 5 min and filtered. These extracts (4 205
mL) were treated with 0.45 g of potassium bromide, 200 µL of sulfuric acid (10 mL/100 206
mL), and 300 µL of potassium bromate solution (0.1 mol/L). After 1 h in the dark at 4°C, 207
the bromination reaction was terminated by adding of 1 mol/L sodium thiosulfate until 208
solutions became colorless, and solutions were extracted with 5 mL of ethyl acetate/hexane 209
(4:1). Organic layers were recovered after centrifugation at 2000 x g for 10 min, and were 210
dried with sodium sulfate and evaporated to dryness under nitrogen. Each sample was 211
dissolved in 50 µL of ethyl acetate, treated with 25 µL of triethylamine, and analyzed by 212
GC–MS. The ions monitored for the identification of the analyte, 2- bromopropenamide, 213
were [C3H4NO]+ = 70, [C3H479BrNO]+ = 149, and [C3H4
81BrNO]+ = 151, using m/z 149 for 214
quantification. The ions monitored for the identification of the corresponding derivative 2-215
bromo[2H2]propenamide were [C22H2H
81Br]+ = 110 and [C32H2H2
81BrNO]+ = 153, using 216
m/z 153 for quantification. 217
GC–MS analyses were conducted with a Perkin Elmer GC Clarus 500 coupled with a 218
Perkin Elmer Clarus 560 MSD (Mass Selective Detector-Quadrupole type). In most 219
experiments, a 30 m × 0.32 mm i.d. × 0.25 µm Elite-5MS capillary column was used. 220
Working conditions were as follows: carrier gas helium (1 mL/min at constant flow); 221
12
injector, 250°C; oven temperature: from 50 (10 min) to 240°C at 5°C/min and then to 222
300°C at 10°C/min; transfer line to MSD, 280°C; ionization EI, 70 eV. 223
Quantification of acrylamide was carried out by preparing standard curves of this 224
compound. Acrylamide content was directly proportional to the acrylamide/internal 225
standard area ratio (r = 0.999, p < 0.0001). The coefficients of variation at the different 226
concentrations were lower than 10%. 227
2.9. Statistical analysis 228
All results were expressed as mean± SD values (n=3). When significant F values were 229
obtained, group differences were evaluated by the Tukey test. All statistical procedures 230
were carried out using the JMP 9.0 package (SAS Institute Inc., Cary, NC). The 231
significance level was p < 0.05 unless otherwise indicated. 232
3. Results and discussion 233
Figure 1 shows the desorption isotherms of the tortilla used to make tortilla chips at 25, 30 234
and 35 °C. Sigmoidal isotherms described as type by Brunauer et al. (1940) were 235
determined. Since desorption is an endothermic process, the increment of the temperature at 236
constant aw enhances the moisture desorbed from the tortilla. Similar desorption isotherms 237
have been reported for starchy products such as potato (McLaughlin & Magee, 1998), 238
maize (Samapundo et al. 2006) and banana (Yan et al. 2008), chestnut flour (Chenlo et al. 239
2011) and Japanese noodle (Inazu et al. 2001). Table 1 gives the estimated parameters 240
obtained by fitting the GAB equation to experimental data. As can be seen from the table, 241
the GAB equation describes adequately the experimental data over the whole measured 242
range of aw. Thus the values of r2 were very close to unit and the P value was ≤ 10 % under 243
13
the studied conditions, which it indicates a good fit (Lomauro et al. 1985). As expected, 244
maximum water desorption by strongly binding sites (monolayer) predicted for GAB 245
equation decreased with increasing temperature. The moisture content in the monolayer 246
(Mo) was 10.12, 9.46 and 8.96 g water/100 g dry solid (corresponding to aw of 0.32, 0.31 247
and 0.30) for 25, 30 and 35 °C, respectively. It is assumed that the value of Mo is a 248
parameter for estimating the amount of water bound to specific polar sites in dehydrated 249
food and in this water content; a storage product should be stable against degradation 250
reactions (Rahman & Labuza, 1999). The constant C is a measure of the attraction force 251
between the water and the sorption sites. The values obtained are higher than those reported 252
by Polou et al. (1997) in commercial tortilla chips. In this study, the k values increased with 253
increasing temperature. The value of k provides a measure of the interaction of the 254
molecules in the multilayers with the adsorbent, and tends to fall between the energy value 255
of the molecules in the monolayer and the liquid water. When k = 1, the properties of the 256
water present in the multilayer are similar to those of the free water (Perez-Alonso et al. 257
2006). 258
Variations in the differential and integral entropy with respect to aw of the tortilla used to 259
make tortilla chips are represented in Figure 2. The intersection of the curves is found in 260
the water content and aw corresponding to the minimum integral entropy. It could be 261
observed that as tortilla desorbed moisture, their integral entropy fell to a minimum. 262
Integral entropy can be directly related to the order-disorder of water molecules sorbed on 263
food, and therefore it is a useful function to study the effect of drying method on the 264
stability of the product. The minimum integral entropy is expected where strong bonds 265
between the adsorbate and the adsorbent are (Nunes & Rotstein, 1991), and hence water 266
14
molecules are less available to participate in spoilage reactions. This point has been 267
proposed as the most suitable for storage and it is considered as the water content 268
corresponding to the monolayer (Dominguez et al. 2007; Hill et al. 1951). 269
Although the value of minimum entropy may be unique, there are food products with zones 270
in which this minimum does not vary appreciably in a defined range of moisture. As it can 271
be observed in Figure 2, the minimum integral entropy at 30 °C for tortilla corresponds to 272
aw value of 0.53 (13.20 g water/100g dry solids). It should be noted that although the 273
minimum integral entropy occurs in a given aw, this minimum does not vary appreciably in 274
a defined range of water activities (0.45 to 0.62). Differential entropy had a minimum at 7.3 275
g water/100 g dry solids (aw=0.18). This parameter does not mean order or disorder of the 276
total system. The differential entropy represents the algebraic sum of the integral entropy at 277
a particular hydration level, plus the change of order or disorder after new water molecules 278
were desorbed by the system at the same hydration level. If the values of moisture content 279
corresponding to minimum integral entropy and minimum differential entropy are different, 280
this particular hydration level at the minimum differential entropy cannot be considered as 281
the maximum stability point, because not all available active sites have been occupied at 282
that particular water content, and therefore it is possible to obtain after this point lower 283
differential changes that provide a better ordering of the water molecules desorbed on food 284
(Beristain et al. 2002). 285
The minimum integral entropy approach has been verified in several studies (see for 286
example: Beristain et al. 1994; Bonilla et al. 2010; Carrillo -Navas et al. 2011; Dominguez 287
et al. 2007; Rascon et al. 2011). 288
15
On the other hand, frequently the thermodynamic analysis is not in accordance with the 289
monolayer obtained with the GAB model (Beristain & Azuara, 1990; Beristain et al. 2002; 290
Nunes & Rotstein, 1991). In this study, the minimum integral entropy point is higher than 291
the calculated Mo value with GAB equation. Since the BET theory, which assumes that the 292
surface of the adsorbent is energetically uniform and does not take into account horizontal 293
interactions between the molecules in the adsorbed layer (Dollimore et al. 1976), is also 294
present in the GAB equation, the GAB monolayer concept has a limited applicability. 295
In this context, the Maillard reaction (main pathway acrylamide formation pathway) is a 296
reaction that requires the interaction of two reagents. Therefore, those factors which affect 297
the concentration and mobility of the reactants will have a significant impact on the 298
reaction rate (Bell, 2007). Thus, the plasticizing effect of water significantly influences on 299
the rate of the Maillard reaction (Bell et al. 1998; White & Bell, 1999). In the minimum 300
integral entropy point, water is less reactive to carry out spoilage reactions in the food, so 301
the mobility of acrylamide precursors is reduced, affecting it formation during thermal 302
processing. 303
Figure 3 shows acrylamide content in tortilla chips prepared from tortillas stored at 304
different water activities. All water activities induced the formation of acrylamide in the 305
tortilla chips, but tortilla chips made of no-stored tortilla (aw=0.98) always contained less 306
acrylamide (688.55±65 µg/kg) than those made of tortilla stored at different water activities 307
(p < 0.05). This suggests a dilution of acrylamide precursors caused by high moisture 308
content. This effect is similar to that reported for the Maillard reaction in which a reduction 309
in reaction rate is observed in aw close to 1 because of the reagent dilution as well as the 310
inhibitory effect that water exerts in this reaction (Ames, 1990; Acevedo et al. 2008). Thus, 311
16
a reduction in the acrylamide content was observed in the minimum integral entropy point. 312
The acrylamide content of tortilla chips prepared from tortilla stored at aw=0.53 (741.85±88 313
µg/kg) was similar to those prepared from no-stored tortilla (p < 0.05). This point is 314
characterized by strong water-food interactions causing a reduction in the diffusion and 315
mobility of the system and affecting to the development of chemical reactions during frying 316
such as the acrylamide formation. On the other hand, an increase on acrylamide content on 317
tortilla chips proportional to reduction of aw (0.43-0.11) on tortilla was observed. This 318
increase may be linked to the concentration of the precursors as well as the evaporation of 319
water which takes place during the start of frying. At this stage, the temperature of the 320
tortilla does not exceed 100 ˚C until a certain amount of water is evaporated. Since 321
acrylamide is not formed at temperatures below 100 ˚C, an increase on acrylamide 322
formation is observed when frying is carried out at low aw where a rapid increase of the 323
internal temperature of the food occurs (Mestdagh et al. 2006). 324
Foods are heterogeneous and dynamic mixtures of macromolecules, solvent and solutes, 325
therefore, the positive or negative correlation between water activity and acrylamide 326
content are probably due to differences in composition and structure of used food systems. 327
According to Viveros- Contreras et al. (2013) and Vigano et al. (2011), food with the same 328
chemical composition and different microstructure could show different moisture sorption 329
behavior, and therefore the aw where spoilage reactions occurs, could be modified . This 330
fact suggests that the microstructure and physical structure of the water in the food is able 331
to restrict the mobility of the precursors responsible for the formation of acrylamide to a 332
specific humidity level during thermal processing and it explains the decrease of the 333
acrylamide content of tortilla elaborated from tortillas equilibrated at the aw corresponding 334
17
to minimum integral entropy zone. Thus, the influence of aw on the acrylamide formation 335
will be different depending on each particular food system. 336
Texture and color are considered the most important parameters of quality and acceptability 337
of fried products. Figure 4 shows the effect of tortilla water activity on the parameters of 338
tortilla chips above mentioned. The color has been correlated with the acrylamide 339
generation in thermally processed foods (Gökmen & Şenyuva, 2006; Lukac et al. 2007; 340
Majcher & Jelen, 2007; Pedreschi et al. 2007). In this study, the appearance of the tortilla 341
chips (Figure 4a) remained constant in the range of aw from 0.43 to 0.84 and increased 342
proportionally to aw reduction (0.43-0.11). It should be emphasized that this change was 343
observed in the water activity zone corresponding to the minimum integral entropy. The 344
fracture force (Figure 4b) showed similar values over the whole range of water activities 345
studied. The obtained fracture force values are close to those found (8.73–9.63 N) by Tseng 346
et al. (1996), and higher than those (5.33–6.17 N) described in commercial tortilla chips 347
(Lujan-Acosta & Moreira, 1997). 348
4. Conclusions 349
The effect of aw on the acrylamide content of tortilla chips could be related to the moisture 350
sorption behavior. The minimum integral entropy in conjunction with the corresponding aw 351
obtained from the desorption isotherm provided a reliable indicator to establish the most 352
appropriate moisture conditions to obtain tortilla chips with reduced level of acrylamide 353
when tortilla was dehydrated. Furthermore, the acrylamide content reduction was similar 354
when tortilla chips were prepared from non-stored tortilla with a moisture content 355
corresponding to water activities close to 1. It should be noted that a partial tortilla 356
18
dehydration to a water activity different to the minimum integral entropy results in tortilla 357
chips with higher acrylamide content. Although a detailed sensory evaluation of tortilla 358
chips prepared is needed, preliminary results showed that the products obtained from 359
tortilla stored in the minimum integral entropy aw did not show any significant change in 360
the appearance of the food product. Therefore, the results suggest that the control of aw, 361
moisture content and the physical state of water in the tortillas are factors to reduce the 362
formation of acrylamide during processing of tortilla chips. This is of the upmost 363
importance, because it provides the food manufacturer with information of which aw is 364
more likely to get a low acrylamide content product over a wider range of processing 365
conditions. 366
Acknowledgments 367
We are indebted to Juan Veles, Edmundo Gutierrez, Carlos Alberto Ávila, Araceli 368
Mauricio and Veronica Flores from CINVESTAV Querétaro for their technical assistance. 369
References 370
AACC International (2010). Approved Method 02-52.01. Approved Methods of the 371
American Association of Cereal Chemists. (11 th Ed.). St Paul, MN, USA. 372
Acevedo, N. C., Briones, V., Buera, P., & Aguilera, J. M. (2008). Microstructure affects the 373
rate of chemical, physical and color changes during storage of dried apple discs. Journal of 374
Food Engineering, 85(2), 222-231. 375
Ames, J. M. (1990). Control of the Maillard reaction in food systems. Trends in Food 376
Science & Technology, 1, 150-154. 377
19
Andrawes, F., Greenhouse, S., & Draney, D. (1987). Chemistry of acrylamide bromination 378
for trace analysis by gas chromatography and gas chromatography—mass spectrometry. 379
Journal of Chromatography A, 399, 269-275. 380
Azuara, E., Vernon-Carter, E. J., & Beristain, C. I. (2007). Thermodynamic analysis of the 381
effect of water activity on the stability of macadamia nut. Journal of Food Engineering, 382
81(3), 566-571. 383
Bassama, J., Brat, P., Bohuon, P., Hocine, B., Boulanger, R., & Günata, Z. (2011). 384
Acrylamide kinetic in plantain during heating process: Precursors and effect of water 385
activity. Food Research International, 44(5), 1452-1458. 386
Bell, L. N. (2007). Moisture effects on food’s chemical stability. In G. V. Barbosa- 387
Canovas, A. J. J. Fontana, S. J. Schmidt, & T. P. Labuza (Eds.), Water activity in foods: 388
Fundamentals and applications (pp. 177–198). Ames: Blackwell Publishing Ltd. 389
Bell, L. N., Touma, E., White, K. L., & Chen, Y. (1998). Glycine loss and maillard 390
browning as related to the glass transition in a model food system. Journal of Food Science, 391
63(4), 625–628. 392
Beristain, C., & Azuara, E. (1990). Estabilidad máxima en productos deshidratados. 393
Ciencia, 41(3), 229–236. 394
Beristain, C. I., Diaz, R., Garcia, H. S., & Azuara, E. (1994). Thermodynamic behavior of 395
green whole and decaffeinated coffee beans during adsorption. Drying Technology, 12(5), 396
1221-1233. 397
20
Beristain, C. I., Azuara, E., & Vernon‐Carter, E. J. (2002). Effect of Water Activity on the 398
Stability to Oxidation of Spray‐Dried Encapsulated Orange Peel Oil Using Mesquite Gum 399
(Prosopis Juliflora) as Wall Material. Journal of Food Science, 67(1), 206-211. 400
Bonilla, E., Azuara, E., Beristain, C. I., & Vernon-Carter, E. J. (2010). Predicting suitable 401
storage conditions for spray-dried microcapsules formed with different biopolymer 402
matrices. Food Hydrocolloids, 24(6), 633-640. 403
Brunauer, S., Deming, L. S., Deming, W. E., & Teller, E. (1940). On a theory of the van 404
der Waals adsorption of gases. Journal of the American Chemical Society, 62(7), 1723-405
1732. 406
Carrillo-Navas, H., González-Rodea, D. A., Cruz-Olivares, J., Barrera-Pichardo, J. F., 407
Román-Guerrero, A., & Pérez-Alonso, C. (2011). Storage stability and physicochemical 408
properties of passion fruit juice microcapsules by spray-drying. Revista Mexicana de 409
Ingeniería Química, 10(3), 421-430. 410
Castle, L., Campos, M. J., & Gilbert, J. (1991). Determination of acrylamide monomer in 411
hydroponically grown tomato fruits by capillary gas chromatography—mass spectrometry. 412
Journal of the Science of Food and Agriculture, 54(4), 549-555. 413
Chenlo, F., Moreira, R., Prieto, D. M., & Torres, M. D. (2011). Desorption isotherms and 414
net isosteric heat of chestnut flour and starch. Food and Bioprocess Technology, 4(8), 415
1497-1504. 416
21
De Vleeschouwer, K., Plancken, I. V. D., Van Loey, A., & Hendrickx, M. E. (2007). 417
Kinetics of acrylamide formation/elimination reactions as affected by water activity. 418
Biotechnology Progress, 23(3), 722-728. 419
De Vleeschouwer, K., Van der Plancken, I., Van Loey, A., & Hendrickx, M. E. (2008). 420
Investigation of the influence of different moisture levels on acrylamide 421
formation/elimination reactions using multiresponse analysis. Journal of Agricultural and 422
Food Chemistry, 56(15), 6460-6470. 423
Dollimore, D., Spooner, P., & Turner, A. (1976). The BET method of analysis of gas 424
adsorption data and its relevance to the calculation of surface areas. Surface Technology, 425
4(2), 121-160. 426
Gökmen, V., & Şenyuva, H. Z. (2006). Study of colour and acrylamide formation in coffee, 427
wheat flour and potato chips during heating. Food Chemistry, 99(2), 238-243. 428
Hedegaard, R. V., Frandsen, H., Granby, K., Apostolopoulou, A., & Skibsted, L. H. (2007). 429
Model studies on acrylamide generation from glucose/asparagine in aqueous glycerol. 430
Journal of Agricultural and Food Chemistry, 55(2), 486-492. 431
Hidalgo, F. J., Delgado, R. M., & Zamora, R. (2009). Degradation of asparagine to 432
acrylamide by carbonyl-amine reactions initiated by alkadienals. Food Chemistry, 116(3), 433
779–784. 434
Hill, T. L., Emmett, P. H., & Joyner, L. G. (1951). Calculation of Thermodynamic 435
Functions of Adsorbed Molecules from Adsorption Isotherm Measurements: Nitrogen on 436
Graphon1, 2. Journal of the American Chemical Society, 73(11), 5102-5107. 437
22
Hunter, R. S. (1973). The measurement of appearance. Fairfax, VA: 383 Hunter Associates 438
Laboratory. 439
IARC. (1994). IARC monographs on the evaluation of carcinogenic risks to humans (Vol. 440
60). International Agency for Research on Cancer. Lyon, France 441
Inazu, T., Iwasaki, K. I., & Furuta, T. (2001). Desorption isotherms for Japanese noodle 442
(udon). Drying Technology, 19(7), 1375-1384. 443
Labuza, T. P., Kaanane, A., & Chen, J. Y. (1985). Effect of temperature on the moisture 444
sorption isotherms and water activity shift of two dehydrated foods. Journal of Food 445
Science, 50(2), 385-392. 446
Labuza, T. P., McNally, L., Gallagher, D., Hawkes, J., & Hurtado, F. (1972). Stability of 447
intermediate moisture foods. 1. Lipid oxidation. Journal of Food Science, 37(1), 154-159. 448
Lomauro, C. J., Bakshi, A. S., & Labuza, T. P. (1985). Evaluation of food moisture 449
sorption isotherm equations. Part I: Fruit, vegetable and meat products. Lebensmittel-450
Wissenschaft und-Technologie, 18(2), 111-117. 451
Lujan-Acosta, J., & Moreira, R. G. (1997). Effects of different drying processes on oil 452
absorption and microstructure of tortilla chips. Cereal Chemistry, 74(3), 216-223. 453
Majcher, M. A., & Jeleń, H. H. (2007). Acrylamide formation in low-fat potato snacks and 454
its correlation with colour development. Food Additives and Contaminants, 24(4), 337-342. 455
Mathlouthi, M. (2001). Water content, water activity, water structure and the stability of 456
foodstuffs. Food Control, 12(7), 409-417. 457
23
McLaughlin, C. P., & Magee, T. R. A. (1998). The determination of sorption isotherm and 458
the isosteric heats of sorption for potatoes. Journal of Food Engineering, 35(3), 267-280. 459
Mestdagh, F., De Meulenaer, B., Cucu, T., & Van Peteghem, C. (2006). Role of water upon 460
the formation of acrylamide in a potato model system. Journal of Agricultural and Food 461
Chemistry, 54(24), 9092-9098. 462
Mottram, D. S., Wedzicha, B. L., & Dodson, A. T. (2002). Acrylamide is formed in the 463
Maillard reaction. Nature, 419(6906), 448-449. 464
Nunes, R. V., & Rotstein, E. (1991). Thermodynamics of the water-foodstuff equilibrium. 465
Drying Technology, 9(1), 113-137. 466
Othmer, D. F. (1940). Correlating vapor pressure and latent heat data. A new plot. 467
Industrial and Engineering Chemistry, 32(6), 841–856. 468
Palou, E., Lopez-Malo, A., & Argaiz, A. (1997). Effect of temperature on the moisture 469
sorption isotherms of some cookies and corn snacks. Journal of Food Engineering, 31(1), 470
85-93. 471
Pascual-Pineda, L. A., Flores-Andrade, E., Alamilla-Beltrán, L., Chanona-Pérez, J. J., 472
Beristain, C. I., Gutiérrez-López, G. F., & Azuara, E. (2013). Micropores and Their 473
Relationship with Carotenoids Stability: A New Tool to Study Preservation of Solid Foods. 474
Food and Bioprocess Technology, 7(4), 1160-1170. 475
Pedreschi, F., Kaack, K., & Granby, K. (2004). Reduction of acrylamide formation in 476
potato slices during frying. LWT-Food Science and Technology, 37(6), 679-685. 477
24
Pérez-Alonso, C., Beristain, C. I., Lobato-Calleros, C., Rodriguez-Huezo, M. E., & 478
Vernon-Carter, E. J. (2006). Thermodynamic analysis of the sorption isotherms of pure and 479
blended carbohydrate polymers. Journal of Food Engineering, 77(4), 753-760. 480
Quirijns, E. J., Van Boxtel, A. J., van Loon, W. K., & Van Straten, G. (2005). Sorption 481
isotherms, GAB parameters and isosteric heat of sorption. Journal of the Science of Food 482
and Agriculture, 85(11), 1805-1814. 483
Rahman, M. S., & Labuza, T. P. (1999). Water activity and food preservation. In M. S. 484
Rahman (Ed.), Handbook of food preservation (pp. 339-382). New York: Marcel Dekker. 485
Rascón, M. P., Beristain, C. I., García, H. S., & Salgado, M. A. (2011). Carotenoid 486
retention and storage stability of spray-dried encapsulated paprika oleoresin using gum 487
Arabic and soy protein isolate as wall materials. LWT-Food Science and Technology, 44 488
(2), 549-557. 489
Rizvi, S. S. H., & Benado, A. L. (1984). Thermodynamics properties of dehydrated food. 490
Food Technology, 38(3), 83–92. 491
Salazar, R., Arámbula-Villa, G., Hidalgo, F. J., & Zamora, R. (2012). Mitigating effect of 492
piquin pepper (Capsicum annuum L. var. Aviculare) oleoresin on acrylamide formation in 493
potato and tortilla chips. LWT-Food Science and Technology, 48(2), 261-267. 494
Samapundo, S., Devlieghere, F., Meulenaer, B. D., Atukwase, A., Lamboni, Y., & 495
Debevere, J. M. (2007). Sorption isotherms and isosteric heats of sorption of whole yellow 496
dent corn. Journal of Food Engineering, 79(1), 168-175. 497
25
Spencer, P. S., & Schaumburg, H. H. (1974). A review on acrylamide neurotoxicity. Part I. 498
Properties, uses and human exposure. Canadian Journal of Neurological Science, 1(2), 499
151–169. 500
Stadler, R. H., Blank, I., Varga, N., Robert, F., Hau, J., Guy, P. A., Robert, M. C., & 501
Riediker, S. (2002). Acrylamide from Maillard reaction products. Nature, 41 (6906), 449– 502
454. 503
Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., & Tornqvist, M. (2002). Analysis of 504
acrylamide, a carcinogen formed in heated foodstuffs. Journal of Agricultural and Food 505
Chemistry, 50 (17), 4998–5006. 506
Tseng, Y. C., Moreira, R., & Sun, X. (1996). Total frying‐use time effects on soybean‐oil 507
deterioration and on tortilla chip quality. International Journal of Food Science & 508
Technology, 31(3), 287-294. 509
Viganó, J., Azuara, E., Telis, V., Beristain, C. I., Jiménez, M., & Telis-Romero, J. (2012). 510
Role of enthalpy and entropy in moisture sorption behavior of pineapple pulp powder 511
produced by different drying methods. Thermochimica Acta, 528, 63-71. 512
Viveros-Contreras, R., Téllez-Medina, D.I., Perea-Flores, M.J., Alamilla-Beltrán, L., 513
Cornejo-Mazón, M., Beristain-Guevara, C. I., Azuara-Nieto E., & Gutiérrez-López, G. F. 514
(2013). Encapsulation of ascorbic acid into calcium alginate matrices through coacervation 515
coupled to freeze-drying. Revista Mexicana de Ingeniería Química, 12(1), 29-39. 516
26
Wexler, A. (1976). Vapor pressure formulation for water in range 0 to 100 °C. A review. 517
Journal of Research of the National Bureau of Standards A Physics and Chemistry, 80(5), 518
775–785 519
White, K. L., & Bell, L. N. (1999). Glucose loss and Maillard browning in solids as 520
affected by porosity and collapse. Journal of Food Science, 6 (6), 1010-1014. 521
Yan, Z., Sousa-Gallagher, M. J., & Oliveira, F. A. (2008). Sorption isotherms and moisture 522
sorption hysteresis of intermediate moisture content banana. Journal of Food Engineering, 523
86 (3), 342-348. 524
Table 1
Estimated GAB parameters for tortilla desorption isotherms
25°C 30°C 35°C
Mo (g H2O/100 g d.s.) 10.11 9.46 8.96
C 14.67 15.72 14.26
k 0.66 0.67 0.70
r2 0.99 0.99 0.99
P (%) 0.96 1.36 1.58
Table 1
Figure captions
Figure 1. Moisture desorption isotherms of tortilla used to make tortilla chips at 25 (), 30
() and 35 (∆) °C. The continuous lines are the fitted points using the GAB model.
Figure 2. Differential () and integral () entropy changes as a function of water activity
for tortilla at 30 °C.
Figure 3. Effect of tortilla water activity on acrylamide content in tortilla chips fried at
180°C for 25 s.
Figure 4. Effect of tortilla water activity on: color (a) and fracture force (b) in tortilla chips
fried at 180°C for 25 s.
Figure Captions
0
5
10
15
20
25
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
aw (-)
Mo
istu
re c
on
ten
t (g
H2O
/10
0 g
d.s
.)
Fig 1. Moisture desorption isotherms of tortilla used to make tortilla chips at 25 (), 30 ()
and 35 (∆) °C. The continuous lines are the fitted points using the GAB model.
Figure 1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0-60
-40
-20
0
20
40
En
tro
py
(J/
mo
l K
)
aw(-)
Fig 2. Differential () and integral () entropy changes as a function of water activity for
tortilla at 30 °C.
Figure 2
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00
300
600
900
1200
1500
Acr
yla
mid
e (
g/k
g)
aw(-)
Fig 3. Effect of tortilla water activity on acrylamide content in tortilla chips fried at 180°C
for 25 s.
Figure 3
42
48
54
60
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00
3
6
9
E (
-)
b
a
w(-)
Fra
ctu
re f
orc
e (
N)
a
Fig 4. Effect of tortilla water activity on: color (a) and fracture force (b) in tortilla chips
fried at 180°C for 25 s.
Figure 4